Given all the negative feedback on the original wind rule... yes, I think we need a new one!
I will make an option on game creation so that it's possible to play with a new rule or using the original one.
As for the two problems noted by Electro:
1) Wind changes too often.
Every scenario has a wind change number. Every three turns, a dice is rolled: if the result is equal or higher than the wind change number, a further roll will be made on both the "Wind velocity change table" and on the "Wind direction change table".
You can see both tables here:
http://www.youplay.it/beta/play/ws_adv_tables.asp(These tables also include "no change" as a possible result).
So, if wind changes too often in a scenario...well, it's down to that number.
If it's a scenario we made here, we can change it.
If it's an original scenario.. we could change it but the scenario was probably meant to be that way, so we'll leave it as is.
One further thing that we could do on this matter is to make the "wind change number" of a scenario explicit (it is not shown anywhere currently), so that people know what to expect from a scenario.
By having the wind change number, a player could also calculate his own forecast with all the percentages as Tomslade suggested.
2) The direction is too random.
Indeed, when the wind changes, it can change a lot. The 120 degrees changes probably happen too often and everytime the wind changes a "new game" starts each time.
There is actually one result in the Wind Direction table that brings the wind back to the original initial direction: this is good as the wind should have a "main direction" and change from that only temporarily to be somewhat realistic (or just to make it more predictable).
However, the "back to initial direction" result has very low probability (1/6 in the table) so instead of giving some "sanity" to this rules, it actually worsens them as it means that sometimes wind will change by 180 degrees!
I'm now trying to make up a new "Wind direction change table".
This table will be created on the core ideas that wind should:
1) Change only by 60° degrees most of the time
2) Usually try to move back to the original direction
3) The few times wind will change by 120° degrees, it will be towards the original direction
4) Never change by 180 degrees in a single turn
I'll run some computer simulations to "validate" the new table before we actually try and play it.
These simulations should confirm by numbers that the new table behaves properly: for example it will give us the exact percentage of how often we can expect a 120° shift (which should be pretty low) or how long and how much the wind can stay away from the original direction (not much/long usually).
Example of how the table could be:
http://i2.minus.com/ihPh2MEIqcRKC.png
